Test for completely broken HTML.

git-svn-id: file:///home/phd/archive/SVN/bookmarks_db/trunk@282 fdd5c36f-1aea-0310-aeeb-c58d7e2b6c23

diff --git a/Robots/parse_html_lxml.py b/Robots/parse_html_lxml.py
index 7aba09cbf5d426e2edf21c96fc4ae3375cef36ac..638af4e972cab376c1fa2f142494a8eea74b9cce 100644 (file)
--- a/Robots/parse_html_lxml.py
+++ b/Robots/parse_html_lxml.py
@@ -11,6 +11,9 @@ from parse_html_util import HTMLParser
 def parse_html(filename, charset=None, log=None):
     html_tree = parse(filename)
 
+    if t.getroot() is None:
+        return None
+
     title = html_tree.findtext('head/title')
     if title is None:
         title = html_tree.findtext('title')